upstream-archive/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-10-02 19:07:25 -05:00
Code Issues Packages Projects Releases Wiki Activity
Files
b11264
llama.cpp/ggml/src/ggml-cuda
T
History
uvos c2a9e16068 HIP: fix template skip for DKQ > 256 mfma kernels (#29559)
2026-09-28 13:47:29 +02:00
..
template-instances
llama : add llama_prec_policy + model-driven W4A4 path (#24364)
2026-09-25 11:36:35 +03:00
vendors
HIP: bump HIP_VERSION requried for fp8 to avoid missing __hip_fp8_e4m3 support in 6.2 (#29231)
2026-09-25 10:36:36 +03:00
acc.cu
llama/ggml: add LLM training support (#10544)
2025-05-12 14:44:49 +02:00
acc.cuh
…
add-id.cu
musa: fix build warnings (#15258)
2025-08-20 10:17:37 +08:00
add-id.cuh
llama : add gpt-oss (#15091)
2025-08-05 22:10:36 +03:00
allreduce.cu
HIP: Enable AllReduce for ROCm (#27825)
2026-09-15 20:57:41 +02:00
allreduce.cuh
HIP: Enable AllReduce for ROCm (#27825)
2026-09-15 20:57:41 +02:00
arange.cu
…
arange.cuh
…
argmax.cu
ggml : use WARP_SIZE/2 for argmax reduction offset (#18092)
2025-12-17 11:47:01 +08:00
argmax.cuh
ggml/ex: calculate accuracy in graph, adapt MNIST (ggml/980)
2024-10-03 21:17:26 +03:00
argsort.cu
cuda : fix CUB argsort corruption caused by in-place keys (#28389)
2026-09-19 07:32:52 +02:00
argsort.cuh
ggml : process data in smaller chunks in CUDA ggml_top_k() and ggml_argsort() to reduce temporary buffers memory usage (#24776)
2026-07-09 20:07:12 +02:00
binbcast.cu
ggml : address integer overflows in binary ops CUDA implementation (#24706)
2026-06-25 10:06:44 +02:00
binbcast.cuh
CUDA: fuse muls (#21665)
2026-04-10 10:24:09 +08:00
clamp.cu
cuda: unary ops as float + de-duplicate (ggml/1130)
2025-03-03 18:18:11 +02:00
clamp.cuh
…
CMakeLists.txt
Enables Windows ARM64 build with MSVC cl.exe (#28362)
2026-09-28 10:07:27 +02:00
col2im-1d.cu
Ggml/cuda col2im 1d (#24417)
2026-06-18 22:23:01 +02:00
col2im-1d.cuh
Ggml/cuda col2im 1d (#24417)
2026-06-18 22:23:01 +02:00
common.cuh
musa: fix PH1 (MTT S5000) operator failures and build issues (#29193)
2026-09-25 10:16:29 +03:00
concat.cu
cuda : relax tensor contiguity requirements for quantized concat (#25678)
2026-07-15 13:36:32 +02:00
concat.cuh
…
conv2d-dw.cu
cuda : add F16 kernel support for CONV_2D_DW (#29064)
2026-09-24 18:38:46 +02:00
conv2d-dw.cuh
CUDA: add conv_2d_dw (#14265)
2025-06-20 09:50:24 +08:00
conv2d-transpose.cu
CUDA & CPU: support F32 kernel type for CONV_TRANSPOSE_2D (#17094)
2026-03-26 10:19:14 +08:00
conv2d-transpose.cuh
CUDA & CPU: support F32 kernel type for CONV_TRANSPOSE_2D (#17094)
2026-03-26 10:19:14 +08:00
conv2d.cu
cuda : accelerate conv2d with implicit GEMM (#29135)
2026-09-21 23:11:43 +08:00
conv2d.cuh
CUDA: add conv2d (#15635)
2025-08-28 20:33:03 +02:00
conv3d.cu
cuda : add conv3d with implicit GEMM (#29137)
2026-09-24 10:24:57 +03:00
conv3d.cuh
cuda : add conv3d with implicit GEMM (#29137)
2026-09-24 10:24:57 +03:00
conv-transpose-1d.cu
ggml-cuda: optimize conv_transpose_1d indexing (#25310)
2026-07-06 11:49:06 +08:00
conv-transpose-1d.cuh
feat: cuda implementation for ggml_conv_transpose_1d (ggml/854)
2024-07-08 12:23:00 +03:00
convert.cu
ggml-cuda : convert contiguous tensors four elements at a time (#29155)
2026-09-21 18:00:51 +02:00
convert.cuh
CUDA: fix BF16 FA compilation (#20865)
2026-03-22 17:53:33 +01:00
count-equal.cu
ggml: fix zero division in ‘dne’ calculation in CUDA COUNT_EQUAL operator when ‘ne’ is small (#10213)
2024-11-09 08:35:46 +01:00
count-equal.cuh
ggml/ex: calculate accuracy in graph, adapt MNIST (ggml/980)
2024-10-03 21:17:26 +03:00
cp-async.cuh
CUDA: FA support for Deepseek (Ampere or newer) (#13306)
2025-05-09 13:34:58 +02:00
cpy-utils.cuh
cuda : support non-contiguous i32 to i32 copy (#17326)
2025-11-23 11:13:34 +01:00
cpy.cu
cuda : enable i16 and i32 for DUP (#28897)
2026-09-15 11:42:21 +08:00
cpy.cuh
cuda : remove legacy copy-op pointer indirection code (#16485)
2025-10-14 11:53:49 +02:00
cross-entropy-loss.cu
CUDA: add dynamic shared mem to softmax, refactor general usage (#14497)
2025-07-03 07:45:11 +08:00
cross-entropy-loss.cuh
ggml/examples: add backend support for numerical optimization (ggml/949)
2024-09-20 21:15:05 +03:00
cumsum.cu
sampling : add support for backend sampling (#17004)
2026-01-04 22:22:16 +02:00
cumsum.cuh
Add support for CUMSUM and TRI for CUDA. (#17584)
2025-12-04 22:19:51 +01:00
dequantize.cuh
CUDA: add Q2_0 support (#25707)
2026-07-30 12:33:25 +03:00
diag.cu
Add DIAG for CUDA (#17873)
2025-12-09 20:28:57 +01:00
diag.cuh
Add DIAG for CUDA (#17873)
2025-12-09 20:28:57 +01:00
diagmask.cu
…
diagmask.cuh
…
dsv4-hc.cu
qwen4exp: add hc ops (#28901)
2026-09-16 16:00:01 +08:00
dsv4-hc.cuh
DeepseekV4: Add fused hyper-connection ops (#25585)
2026-07-17 00:33:33 +08:00
fattn-common.cuh
CUDA: tune FA for Gemma 4 on Ampere or newer (#29152)
2026-09-20 22:20:12 +02:00
fattn-mma-f16.cuh
HIP: fix template skip for DKQ > 256 mfma kernels (#29559)
2026-09-28 13:47:29 +02:00
fattn-tile.cu
HIP: remove rocWMMA FlashAttention (#26046)
2026-07-24 17:53:54 +02:00
fattn-tile.cuh
CUDA: tune fp16 tile FlashAttention configs for head sizes 40-112 (#26289)
2026-09-27 13:07:37 +03:00
fattn-vec.cuh
cuda: fixes races in mmid and mmf (#28475)
2026-09-06 19:45:01 +08:00
fattn.cu
HIP: Enable fattn-mma kernel on cdna for dkq > 256 for large batch sizes (#28907)
2026-09-27 13:06:56 +03:00
fattn.cuh
cuda: reserve space for quantize kv-cache at startup (#23907)
2026-06-03 18:39:59 +08:00
fill.cu
ggml : allow fill node alloc inplace (#17870)
2025-12-09 12:23:47 +01:00
fill.cuh
cuda : add FILL op support (#17851)
2025-12-08 21:10:12 +08:00
fwht.cu
cuda: add F16 input to the FWHT (#29096)
2026-09-26 21:36:09 +02:00
fwht.cuh
cuda: add F16 input to the FWHT (#29096)
2026-09-26 21:36:09 +02:00
gated_delta_net.cu
Remove redundant CUDA copies after gated_delta_net. (#23940)
2026-07-03 14:36:29 +05:30
gated_delta_net.cuh
Remove redundant CUDA copies after gated_delta_net. (#23940)
2026-07-03 14:36:29 +05:30
getrows.cu
CUDA: add Q2_0 support (#25707)
2026-07-30 12:33:25 +03:00
getrows.cuh
CUDA: batched+noncont MMQ, refactor bs>1 MoE code (#13199)
2025-04-30 23:12:59 +02:00
ggml-cuda.cu
cuda: support Nemotron 3 Puzzle state size 96 for ssm scan (#28717)
2026-09-26 22:05:52 +02:00
gla.cu
llama: add support for QRWKV6 model architecture (#11001)
2025-01-10 09:58:08 +08:00
gla.cuh
llama: add support for QRWKV6 model architecture (#11001)
2025-01-10 09:58:08 +08:00
im2col.cu
CUDA/HIP: improve access patterns in im2col (#28013)
2026-09-16 13:46:21 +02:00
im2col.cuh
ggml: add ops for WAN video model (cuda && cpu) (#15669)
2025-09-04 10:38:49 +02:00
lightning-indexer.cu
cuda : CUDA GGML_OP_LIGHTNING_INDEXER implementation (generic vector kernel + wmma kernel) (#25545)
2026-07-15 19:57:52 +02:00
lightning-indexer.cuh
cuda : CUDA GGML_OP_LIGHTNING_INDEXER implementation (generic vector kernel + wmma kernel) (#25545)
2026-07-15 19:57:52 +02:00
mean.cu
cuda: support row-contiguous SUM_ROWS (#26308)
2026-09-15 12:39:29 +02:00
mean.cuh
CUDA: add mean operation (#14313)
2025-06-22 12:39:54 +08:00
mma.cuh
cuda: fix sm_70 tile compilation error (#29224)
2026-09-21 19:11:29 +03:00
mmf.cu
CUDA: dedup MoE gate/up activation quantization (#25441)
2026-07-16 09:02:25 +02:00
mmf.cuh
cuda: fixes races in mmid and mmf (#28475)
2026-09-06 19:45:01 +08:00
mmid.cu
cuda: fixes races in mmid and mmf (#28475)
2026-09-06 19:45:01 +08:00
mmid.cuh
CUDA: dedup MoE gate/up activation quantization (#25441)
2026-07-16 09:02:25 +02:00
mmq-config-ampere.cuh
Revert "CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (#24546)" (#28551)
2026-09-07 16:28:19 +02:00
mmq-config-blackwell.cuh
Revert "CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (#24546)" (#28551)
2026-09-07 16:28:19 +02:00
mmq-config-cdna.cuh
Revert "CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (#24546)" (#28551)
2026-09-07 16:28:19 +02:00
mmq-config-gcn.cuh
ggml-cuda: hip add specific config table for AMD GCN (#27841)
2026-09-12 11:26:53 +02:00
mmq-config-pascal-dp4a.cuh
Revert "CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (#24546)" (#28551)
2026-09-07 16:28:19 +02:00
mmq-config-pascal-older.cuh
Revert "CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (#24546)" (#28551)
2026-09-07 16:28:19 +02:00
mmq-config-rdna2.cuh
Revert "CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (#24546)" (#28551)
2026-09-07 16:28:19 +02:00
mmq-config-rdna3-5.cuh
Revert "CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (#24546)" (#28551)
2026-09-07 16:28:19 +02:00
mmq-config-rdna3.cuh
Revert "CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (#24546)" (#28551)
2026-09-07 16:28:19 +02:00
mmq-config-rdna4.cuh
Revert "CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (#24546)" (#28551)
2026-09-07 16:28:19 +02:00
mmq-load-tiles.cuh
llama : add llama_prec_policy + model-driven W4A4 path (#24364)
2026-09-25 11:36:35 +03:00
mmq-vec-dot.cuh
llama : add llama_prec_policy + model-driven W4A4 path (#24364)
2026-09-25 11:36:35 +03:00
mmq.cu
llama : add llama_prec_policy + model-driven W4A4 path (#24364)
2026-09-25 11:36:35 +03:00
mmq.cuh
llama : add llama_prec_policy + model-driven W4A4 path (#24364)
2026-09-25 11:36:35 +03:00
mmvf.cu
ggml: add SWIGLU_CLAMP (#27930)
2026-08-30 23:00:02 +08:00
mmvf.cuh
CUDA: use mmvq for mul-mat-id for small batch sizes (#18958)
2026-02-03 23:31:23 +08:00
mmvq.cu
cuda : tune MMVQ to MMQ crossover for SM70 (Volta) (#28912)
2026-09-21 10:45:31 +03:00
mmvq.cuh
CUDA: route batch>=4 quantized matmul to MMQ on AMD MFMA hardware (#23227)
2026-05-28 14:50:25 +02:00
moe-weighted-reduction.cu
cuda: fuse MoE weighted expert reduction (#25952)
2026-09-01 21:48:47 +02:00
moe-weighted-reduction.cuh
cuda: fuse MoE weighted expert reduction (#25952)
2026-09-01 21:48:47 +02:00
norm.cu
CUDA: fuse RMS_NORM + SCALE into one kernel (#29393)
2026-09-25 08:22:43 +03:00
norm.cuh
CUDA: fuse RMS_NORM + SCALE into one kernel (#29393)
2026-09-25 08:22:43 +03:00
opt-step-adamw.cu
ggml: new optimization interface (ggml/988)
2024-11-17 08:30:29 +02:00
opt-step-adamw.cuh
ggml/examples: add backend support for numerical optimization (ggml/949)
2024-09-20 21:15:05 +03:00
opt-step-sgd.cu
finetune: SGD optimizer, more CLI args (#13873)
2025-08-14 12:03:57 +02:00
opt-step-sgd.cuh
finetune: SGD optimizer, more CLI args (#13873)
2025-08-14 12:03:57 +02:00
out-prod.cu
ggml-cuda: provide static workspace for cuBLAS handles (#26574)
2026-08-20 10:27:51 +03:00
out-prod.cuh
ggml/examples: add backend support for numerical optimization (ggml/949)
2024-09-20 21:15:05 +03:00
pad_reflect_1d.cu
musa: fix build warnings (#15611)
2025-09-26 02:56:10 +02:00
pad_reflect_1d.cuh
cuda : add Pad Reflect 1D support (#14659)
2025-08-22 13:06:29 +02:00
pad.cu
cuda : extend GGML_OP_PAD to work with non-cont src0 (#19429)
2026-02-10 08:07:16 +02:00
pad.cuh
…
pool1d.cu
cuda : add POOL_1D support (#27573)
2026-08-23 10:37:32 +03:00
pool1d.cuh
cuda : add POOL_1D support (#27573)
2026-08-23 10:37:32 +03:00
pool2d.cu
…
pool2d.cuh
…
quantize.cu
cuda: fix warnings for unused variable/function (#26688)
2026-08-07 07:51:56 +03:00
quantize.cuh
CUDA: Improve NVFP4 W4A4 activation quantization (#25730)
2026-07-22 19:28:02 +02:00
reduce_rows.cuh
cuda: support row-contiguous SUM_ROWS (#26308)
2026-09-15 12:39:29 +02:00
roll.cu
CUDA: add roll (#14919)
2025-07-29 14:45:18 +08:00
roll.cuh
CUDA: add roll (#14919)
2025-07-29 14:45:18 +08:00
rope.cu
ggml: add ggml_rope_set_offset (+ metal support) (#27120)
2026-08-19 14:04:57 +02:00
rope.cuh
CUDA: fuse rms_norm + mul + rope (+ view + set_rows) (#26767)
2026-08-09 00:32:37 +08:00
scale.cu
Programmatic Dependent Launch (PDL) for more performance on newer NVIDIA GPUs (Hopper+) (#22522)
2026-05-20 13:59:02 +02:00
scale.cuh
…
set-rows.cu
cuda : add support for f16->f16 GGML_OP_SET_ROWS (#25367)
2026-07-08 19:24:20 +08:00
set-rows.cuh
CUDA: add set rows for f32 and f16 (#14551)
2025-07-12 16:31:38 +03:00
set.cu
cuda: add SET operation support (#16804)
2025-10-28 20:10:28 +01:00
set.cuh
cuda: add SET operation support (#16804)
2025-10-28 20:10:28 +01:00
snake.cu
cuda: fuse snake activation (mul, sin, sqr, mul, add) (#22667)
2026-05-08 17:44:09 +08:00
snake.cuh
cuda: fuse snake activation (mul, sin, sqr, mul, add) (#22667)
2026-05-08 17:44:09 +08:00
softcap.cu
Programmatic Dependent Launch (PDL) for more performance on newer NVIDIA GPUs (Hopper+) (#22522)
2026-05-20 13:59:02 +02:00
softcap.cuh
cuda : add softcap fusion (#14907)
2025-07-29 14:22:03 +02:00
softmax.cu
CUDA: Fix data-races when reusing SMEM in block_reduce (#26385)
2026-08-03 14:22:44 +02:00
softmax.cuh
CUDA: backwards pass for misc. ops, add tests (#11257)
2025-01-16 16:43:38 +01:00
solve_tri.cu
ggml-cuda: provide static workspace for cuBLAS handles (#26574)
2026-08-20 10:27:51 +03:00
solve_tri.cuh
SOLVE_TRI CUDA kernel for small matrices (#17457)
2025-11-28 12:15:32 +08:00
ssm-conv.cu
Avoid PDL race conditions by disabling __restrict__ when PDL is used (#24030)
2026-06-03 13:56:42 +02:00
ssm-conv.cuh
CUDA: fuse SSM_CONV + ADD(bias) + SILU (#22478)
2026-04-30 02:39:56 +08:00
ssm-scan.cu
cuda: support Nemotron 3 Puzzle state size 96 for ssm scan (#28717)
2026-09-26 22:05:52 +02:00
ssm-scan.cuh
ggml : faster ssm scan (#10558)
2025-03-31 18:05:13 +02:00
sum.cu
CUDA: Optimize reduce_rows_f32 kernel, leading up to 25x perf improvement on kernel-level and 10% perf increase for Gemma3n (#15132)
2025-08-13 10:04:46 +02:00
sum.cuh
tests: add gradient tests for all backends (ggml/932)
2024-09-08 11:05:55 +03:00
sumrows.cu
cuda: support row-contiguous SUM_ROWS (#26308)
2026-09-15 12:39:29 +02:00
sumrows.cuh
CUDA: add mean operation (#14313)
2025-06-22 12:39:54 +08:00
top-k.cu
ROCm: add radix TOP_K for long rows (#27466)
2026-08-31 15:00:04 +02:00
top-k.cuh
sampling : add support for backend sampling (#17004)
2026-01-04 22:22:16 +02:00
topk-moe.cu
musa: fix PH1 (MTT S5000) operator failures and build issues (#29193)
2026-09-25 10:16:29 +03:00
topk-moe.cuh
CUDA: extend MOE fusion to specdec, earlier MOE glu fusion and topk-router fusion were restricted to 1 token (#27621)
2026-08-31 19:22:28 +08:00
tri.cu
Add support for CUMSUM and TRI for CUDA. (#17584)
2025-12-04 22:19:51 +01:00
tri.cuh
Add support for CUMSUM and TRI for CUDA. (#17584)
2025-12-04 22:19:51 +01:00
tsembd.cu
ggml : fix padding in timestep embedding kernels (#15932)
2025-09-16 15:25:57 +02:00
tsembd.cuh
…
unary.cu
ggml: add SWIGLU_CLAMP (#27930)
2026-08-30 23:00:02 +08:00
unary.cuh
ggml: add SWIGLU_CLAMP (#27930)
2026-08-30 23:00:02 +08:00
upscale.cu
model: LFM2-VL fixes (#17577)
2025-11-30 21:57:31 +01:00
upscale.cuh
…
vecdotq.cuh
CUDA: branchless Q4_K/Q5_K unpack to speed up mmvq, L2 prefetch on DGX Spark (#26705)
2026-09-07 19:36:58 +08:00
wkv.cu
musa: fix PH1 (MTT S5000) operator failures and build issues (#29193)
2026-09-25 10:16:29 +03:00
wkv.cuh
llama: Add support for RWKV v7 architecture (#12412)
2025-03-18 07:27:50 +08:00